Part I of this blog summarized Steven Koonin’s criticism of the climate debate: he argues that key scientific findings are quite different from the climate reports headlines – for example, that there is almost no evidence that climate change is causing extreme weather events to occur with greater frequency.
Now let’s get a second opinion: Listen to this podcast with Professor Adam Sobel, a climate science expert at Columbia University who specializes in extreme weather events attribution –figuring out whether hurricanes, heat waves etc. are caused by climate change. I like this podcast because Sobel, who is extremely critical of Koonin, sounds clear and objective in his assessment of the science and the risks, and is very explicit on his view that advocacy for decisive action should be a top priority for scientists.
The criticism goes both ways. Listen to what Koonin thinks of Sobel’s area of specialization: “practitioners argue that event attribution studies are the best climate science can do in terms of connecting weather to changes in climate. But as a physical scientist, I’m appalled that such studies are given credence, much less media coverage.” He quotes the World Meteorological Organization, which says “…any single event, such as a severe tropical cyclone (hurricane or typhoon), cannot be attributed to human-induced climate change, given the current status of scientific understanding.”
Extreme events attribution
And yet, on extreme events Sobel largely agrees with Koonin. He stresses that our understanding of the probable link between climate change and extreme weather events varies considerably depending on the type of event. We are most confident about heat waves: since the globe is getting warmer, heat waves will tend to be hotter. At the other extreme, Sobel says we do not understand whether climate change is likely to lead to more or fewer hurricanes, tornadoes and winter storms.
A few points worth highlighting:
Sobel notes that when we get a heat wave that pushes the maximum temperature to say 45°C, we cannot reject the idea that without climate change we would anyway have gotten a heat wave of 44°C: since the globe has warmed by 1°C, this has raised the baseline, but heat waves would happen anyway.
Have heat waves become more frequent? Sobel says probably yes, but he says this with surprisingly low conviction. Koonin notes that the IPCC fifth assessment (2014) expressed only medium confidence that the length and frequency of heat waves has increased globally since 1950. The sixth assessment (2021) however has upgraded that to high confidence. (The language here is important. Media headlines claim we “know” everything about climate change. The IPCC reports tell us that we “know” nothing; we “think” various things with various degrees of confidence.)
Have high temperatures become more extreme? Sobel implicitly suggests they should have: if the baseline temperature of the globe has risen, the usual swings in temperature should now give higher peaks than ever before. Koonin however provides data showing this is not the case, at least in the US: since 1900 the lowest temperatures in the US have gotten less cold but the highest have not gotten hotter. Koonin shows a chart from Chapter 6 of the USGCRP (US Global Change Research Program) Climate Science Special Report (2017), part of the Fourth National Climate Assessment – reproduced below. The panel on the left shows the coldest temperatures have increased; the panel on the right shows the warmest temperatures have been broadly stable since 1950, at the same level as 1900 and markedly lower than in 1930-40.
Koonin also gives another chart, reproduced below, showing that the number of record high daily temperatures recorded at 725 US weather stations (“record high” means higher than ever recorded before on the same day of the year at the same station) has not increased over the last century; the number of record low temperatures has decreased:
Are you also confused now? We have data showing that high temperature extremes (in the US) have not increased, but the IPCC is highly confident that heat waves have become frequent. Well, what is a heat wave? The latest IPCC report defines it as “a period of abnormally hot weather defined with reference to a relative temperature threshold, lasting from two days to months. Heat waves and warm spells have various and in some cases overlapping definitions.” Various definitions – that’s not very helpful. The Environmental Protection Agency for example defines a heat wave as a period of two or more days when the daily minimum temperature adjusted for humidity is higher than a certain historical average. Ok: Koonin has also shown that minimum temperatures have gotten less cold, and we know that a warmer planet on average brings more humidity. Mystery solved: heat waves defined as periods of warmer than average weather have increased; but it is not true that we are experiencing more frequent bouts of hotter-than-ever temperatures.
By the same token, since global warming has made the coldest temperatures less cold, we certainly cannot argue that it has made extreme cold snaps like the one in Texas more likely. Indeed Sobel dismisses the idea.
Sobel also notes that even taking climate change into account, none of the climate models can account for this summer’s heat wave in the US Pacific Northwest. The heat was so extreme that according to the models it simply should not have happened. This supports Koonin’s point that the models are not reliable; but it also shows that things can get even worse than the models predict, as Sobel observes.
My conclusion: as the planet has gotten warmer, the risk of weather events directly linked to higher temperatures, like heat waves and forest fires, has probably increased; blaming every hurricane and flood on climate change has no scientific basis; the risk of catastrophic weather events might be increasing, or it might not – science does not yet have an answer.
Uncertainty and risk
Here we come to an important point, where Koonin and Sobel exemplify a key disagreement.
Sobel argues that uncertainty is part of the risk we face and should not be an excuse for inaction. We know temperatures have been rising and we know human activity has contributed. True, we do not understand everything that is going on, and the models leave a margin of uncertainty; but that means things could get even worse than the models predict, and therefore we should err on the side of caution and take rapid action.
Koonin, to the contrary, argues that uncertainty counsels a more cautious approach. Yes, temperatures have been rising and human activity has contributed. But our understanding is so limited that we don’t actually know how much human activity has contributed, nor whether the corrective actions we plan will have any meaningful impact compared to the underlying natural changes.
There is no objective right or wrong here, it’s a legitimate disagreement.
Costs and benefits
This matters because reducing emissions quickly carries significant costs. A 2020 study estimates that keeping the global temperature increase below 1.5°C would leave 80 million more people in poverty by 2030 (Figure 2). The recent spike in energy prices shows that reducing reliance on fossil fuels before we make further progress on renewables implies slower growth and lower incomes.
I am enthusiastically optimistic about technological innovation. But here is a reality check: with all the impressive advances in renewable energy technologies, for the last 30 years the share of fossil fuels in total global energy consumption has remained stubbornly above 80%. (Meanwhile global energy consumption has risen by over two-thirds.) In the next 30 years, it must drop to less than 30% to meet the targets of the International Energy Agency’s ‘Net Zero’ scenario. At the current pace of energy transition, we are going nowhere fast.
We need to be realistic and admit that a rapid transition away from fossil fuels involves substantial costs. Advanced economies will have to change their lifestyle, and emerging markets will have to accept a much slower improvement in living standards. Given the substantial costs involved, we need a cost-benefit analysis; and in this analysis, the uncertainty on our climate change forecasts does matter.
Can we trust the data?
Speaking of uncertainty, how confident are we of the “actual” data on global temperature? Before you dismiss me as a lunatic, bear with me a minute:
In 2013, climate scientists observed that global warming had paused. The Economist quoted a NASA scientist noting that “global temperature has been flat for a decade,” illustrated with a chart provided by Professor Ed Hawkins (one of the authors of the IPCC reports), reproduced below:
The Economist was careful to reiterate “that does not mean the problem is going away.” Still, the fact that temperatures had remained stable for over a decade while emissions kept rising was defined as “among the biggest puzzles in climate science just now.”
Two years later the puzzle suddenly disappeared: In 2015 the US National Oceanographic and Atmospheric Administration (NOAA) published an analysis that included ocean temperatures observations made with a new methodology as well as new surface data. According to the amended data, warming had not paused but had continued at the same pace. NOAA concluded: “The IPCC's statement of two years ago - that the global surface temperature 'has shown a much smaller increasing linear trend over the past 15 years than over the past 30 to 60 years' - is no longer valid.”
A conspiracy theorist would tell you that the inconvenient data had been conveniently revised away. For me this just shows that we have significant uncertainty even in current temperature measurements. Note that we are talking about very significant recent revisions to the data of the last 20 years, not to the estimates of 200 years ago. Indeed several scientists reacted with caution to the news, stressing that clearly there is a lot we don’t yet understand.
Think about it for a moment: a few years ago climate scientists told us, ‘finally some encouraging news, global temperatures have stabilized in the last fifteen years, keep up the good work!’; then a couple of years later they said ‘nope, scrap that, we measured it all wrong, temperatures have kept rising.’ Yet at the same time we are asked to believe that precise temperature estimates for the last 2000 years and decimal-point forecasts for the next 200 are indisputable, “settled” science. I don’t know about you, but I am not convinced.
As we factor the uncertainty into our decisions, we have to keep in mind that the margin of uncertainty is so high that we might suddenly discover that recent actual data “are no longer valid.”
Part III of the blog will discuss the social responsibility of scientists and the risks that the current framing of the climate change debate will backfire.